Journal of Neurochemistry, 2005, 92, 925–933
doi:10.1111/j.1471-4159.2004.02939.x
Biochemical and functional evidence for heteromeric assembly of
P2X1 and P2X4 subunits
Annette Nicke,*, ,1 Daniel Kerschensteiner*,1,2 and Florentina Soto*
*Department of Molecular Biology of Neuronal Signals, Max-Planck Institute for Experimental Medicine, Göttingen, Germany
Department of Neurochemistry, Max-Planck Institute for Brain Research, Frankfurt am Main, Germany
Abstract
P2X receptors are ligand-gated ion channels activated by
extracellular ATP. In expression systems, P2X subunits form
homo- and heterotrimeric receptors. Heteromerization is also
likely to occur in vivo as (i) most P2X subunits show overlapping
distribution in different tissues and (ii) the functional properties
of many native P2X receptors differ from those of heterologously expressed homomeric receptors. Here, we used the
Xenopus laevis oocyte expression system to test for heteromerization of P2X1 and P2X4 subunits. Upon co-injection, P2X4
subunits were co-purified with hexahistidyl-tagged P2X1 subunits indicating heteromerization. Blue native polyacrylamide
gel electrophoresis (BN-PAGE) analysis of these P2X complexes excluded artificial aggregation and confirmed that both
subunits were present in trimeric complexes of the same size.
Two-electrode voltage-clamp experiments revealed functional
P2X receptors with kinetic properties resembling homomeric
P2X4 receptors and a pharmacological profile similar to
homomeric P2X1 receptors. Thus, application of a,b-methylene
ATP evoked a slowly desensitizing current sensitive to the
antagonists suramin and 2¢,3¢-O-(2,4,6-trinitrophenyl)-ATP.
This study provides for the first time biochemical and functional
evidence for the formation of heteromeric P2X1+4 receptors.
These receptors may account for native P2X mediated
responses that until now could not be correlated with previously
described recombinant P2X receptors.
Keywords: ATP receptor, blue native polyacrylamide gel
electrophoresis, heterologous expression, nucleotide receptor, purinergic receptor, Xenopus oocytes.
J. Neurochem. (2005) 92, 925–933.
Fast responses to extracellular ATP are mediated by a class of
plasma membrane ligand-gated ion channels called P2X
receptors. P2X receptors are complexes of three P2X subunits,
of which seven different subtypes have been identified in
mammals (P2X1)7) (North 2002). Functional studies of
heterologously expressed homomeric P2X receptors allowed
a classification according to their functional properties in: (i)
rapidly desensitizing, a,b-methylene ATP (abmeATP) sensitive P2X1 and P2X3, (ii) moderately desensitizing abmeATP
insensitive P2X4, and (iii) non-desensitizing, abmeATP
insensitive P2X2, P2X5 and P2X7 receptors (North 2002).
When expressed alone, P2X6 subunits form functional membrane receptors very inefficiently (Collo et al. 1996; Soto et al.
1996a; Le et al. 1998a; King et al. 2000; Jones et al. 2004).
These P2X6 receptors have been described either as nondesensitizing, abmeATP insensitive (Collo et al. 1996) or as
non-desensitizing abmeATP sensitive receptors (Jones et al.
2004). Differences in heteromerization with endogenous P2X
subunits and post-translational modifications have been proposed to account for the discrepancy (Jones et al. 2004).
Additional phenotypes have been described after co-expression of P2X2 and P2X3 (Lewis et al. 1995), P2X1 and P2X5
(Torres et al. 1998; Le et al. 1999; Surprenant et al. 2000),
P2X4 and P2X6 (Le et al. 1998a), P2X2 and P2X6 (King et al.
2000) and P2X1 and P2X2 (Brown et al. 2002) subunits.
Co-immunoprecipitation experiments with tagged P2X
Received June 17, 2004; revised manuscript received October 14, 2004;
accepted October 19, 2004.
Addrress correspondence and reprint requests to Dr Florentina Soto,
Max-Planck-Institut für experimentelle Medizin, Hermann-Rein-Str. 3,
D-37075 Göttingen, Germany. E-mail: fsoto@gwdg.de
1
A. Nicke and D. Kerschensteiner contributed equally to this work.
2
The present address of Daniel Kerschensteiner is Laboratory of
Molecular Pharmacology, Department of Pharmacology, University
College London, Gower Street, London WC1E 6BT, UK.
Abbreviations used: abmeATP, a,b-methylene ATP; BN-PAGE, blue
native polyacrylamide gel electrophoresis; His-P2X1, hexahistidyl-tagged P2X1; Ni2+-NTA, Ni2+-nitrilotriacetic acid; SDS–PAGE, sodium
dodecyl sulfate–polyacrylamide gel electrophoresis; TNP-ATP, 2¢,3¢-O(2,4,6-trinitrophenyl)-ATP.
2005 International Society for Neurochemistry, J. Neurochem. (2005) 92, 925–933
925
926 A. Nicke et al.
subunits pairwise expressed in HEK cells suggest that the
possibilities of heteromerization exceed the number of functional phenotypes described so far (Torres et al. 1999).
Heteromerization of P2X receptor subunits is also likely to
occur in vivo as (i) a number of P2X subunits show overlapping
distribution in different mammalian and non-mammalian
tissues and (ii) many of the in vivo described P2X-receptor
mediated currents show a kinetic and pharmacological behavior that does not correspond to heterologously expressed
homomeric receptors. For instance, the sustained response to
abmeATP observed in sensory neurons of the dorsal root and
nodose ganglia presents pharmacological and kinetic properties that correlate well with the properties of heterologously
expressed heteromeric P2X2+3 receptors (for review see Dunn
et al. 2001). These responses are modified in P2X3 (–/–) mice
(Zhong et al. 2001). However, responses to abmeATP with
pharmacological properties that are different to P2X1, P2X3
and P2X2+3 receptors have been described in neurons of the
trigeminal mesencephalic nucleus (Patel et al. 2001) and in
superior cervical neurons (Calvert and Evans 2004), indicating
the presence of heteromeric P2X receptors of so far unknown
subunit composition.
P2X4 subunits have been detected in most tissues analyzed
so far (Buell et al. 1996; Soto et al. 1996a; Le et al. 1998b;
Bo et al. 2003) where their expression pattern overlaps with
one or more additional P2X subunits (Collo et al. 1996).
Native receptors with functional properties similar to homomeric P2X4 receptors have only been found in the submandibular gland of the rat (Buell et al. 1996), in chicken cardiac
muscle (Hu et al. 2002) and in rabbit osteoclasts (Weidema
et al. 1997; Naemsch et al. 1999). Formation of heteromeric
receptors containing P2X4 subunits could account for the
scarcity of functionally detected native homomeric P2X4
receptors. Until now, functional heteromerization of P2X4
subunits has only been shown with P2X6 subunits (Le et al.
1998a). Here we show that P2X1 and P2X4 subunits form a
novel slowly desensitizing abmeATP sensitive heteromeric
receptor. These findings increase the number of heteromeric
P2X receptor phenotypes described so far and enrich our
understanding of the molecular composition of native P2X
receptors.
Experimental procedures
Synthesis of complementary RNA and injection in defoliculated
Xenopus oocytes
N-terminal hexahistidyl tagged P2X1 (His-P2X1) cDNA in pNKS2
vector was kindly provided by G. Schmalzing. (Nicke et al. 1998).
Rat P2X4 cDNA was subcloned in psGEM vector and capped RNA
was synthethized as previously described (Soto et al. 1996a).
Defoliculated oocytes were injected with 46 nL cRNA per oocyte at
concentrations of 50 ng/lL for P2X1 and P2X4, and 50 or 100 ng/
lL for P2X1+4 (1 : 1 ratio of P2X1 to P2X4 cRNA).
Purification of hexahistidyl-tagged proteins, blue native
polyacrylamide gel electrophoresis and western blot
Protein complexes containing His-tagged P2X subunits were
purified under-non-denaturing conditions from digitonin extracts
of Xenopus laevis oocytes as previously described (Nicke et al.
1998). To prevent unspecific interactions with the Ni2+-nitrilotriacetic acid (Ni2+-NTA) agarose, 30 mM imidazole was included in
the washing buffer. Blue native polyacrylamide gel electrophoresis
(BN-PAGE) was carried out as described (Schägger and Von Jagow
1991; Schägger et al. 1994). Proteins were separated on 4–13%
polyacrylamide gradient gels and subsequently blotted onto polyvinylidene membranes. Sodium dodecyl sulfate–polyacrylamide gel
electrophoresis (SDS–PAGE) separated protein samples were
blotted onto nitrocellulose membranes. Western blotting was
performed as previously described (Rubio and Soto 2001). For the
detection of P2X1 subunits a commercial antibody to P2X1
(Alomone Laboratories, Israel) was used. For detection of P2X4
subunits, a newly developed affinity purified antibody raised as
described (Rubio and Soto 2001) to amino acids 362–388 of the rat
P2X4 subunit was used.
Electrophysiological recordings
Two electrode voltage-clamp recordings were performed 2–5 days
after cRNA injection in Xenopus laevis oocytes. The standard Mg2+
Ringer solution used to superfuse the oocytes contained 115 mM
NaCl, 2.8 mM KCl, 1.8 mM MgCl2, and 10 mM HEPES, pH 7.2
(Mg2+-Ringer). The antagonists 2¢,3¢-O-(2,4,6-trinitrophenyl)-ATP
(TNP-ATP) (Molecular Probes, Eugene, OR, USA) and suramin
(Sigma, München, Germany) were co-applied with ATP during
perfusion with the standard Mg2+ Ringer solution. Voltage and
current electrodes were filled with 2 M KCl solution and had
resistances of 0.4–1.0 MW. All experiments were performed at room
temperature (18–22C). Currents were recorded using a Turbo
TEC-10 CD amplifier (npi electronics, Lambrecht, Germany), low
pass filtered at 100 Hz and sampled at 500 Hz. The oocytes were
voltage clamped at )70 mV. Data are presented as mean ± SE from
n experiments.
Results
P2X4 subunits are co-purified with His-tagged P2X1
subunits in a heterotrimeric complex
To test for an interaction between P2X1 and P2X4 subunits
we co-expressed His-P2X1 subunits with P2X4 subunits in
Xenopus oocytes. Upon solubilization with digitonin, protein
complexes were purified under non-denaturing conditions via
a Ni2+-NTA agarose resin and separated by SDS–PAGE.
Co-purified P2X4 subunits were detected by western blot
using a C-terminal P2X4-specific antibody (Fig. 1a) indicating the formation of heteromeric receptors composed of
P2X1 and P2X4 subunits. No significant immunoreactivity
was detected when extracts from oocytes injected with P2X4
subunits alone (Figs 1a and b) or combined extracts from
oocytes separately injected with His-P2X1 and P2X4 subunits
were subjected to the same procedure (result not shown).
2005 International Society for Neurochemistry, J. Neurochem. (2005) 92, 925–933
Heteromeric receptors formed by P2X1 and P2X4 subunits 927
Fig. 1 Biochemical evidence for heterotrimeric P2X1+4 receptors.
cRNAs encoding the indicated P2X subunits were injected into Xenopus oocytes. At 1–5 days after injection, protein complexes were
purified under non-denaturing conditions via a Ni2+-nitrilotriacetic acid
(Ni-NTA) agarose resin from digitonin (1%) extracts of the oocytes. (a)
Co-purification of the non-tagged P2X4 subunit with the hexahistidyltagged P2X1 (His-P2X1) subunit. Purified protein complexes were
separated by sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS–PAGE) and co-purified P2X4 subunits were detected
by immunoblotting using a P2X4 antibody. (b) Blue native PAGE
(BN-PAGE) analysis of P2X protein complexes. Aliquots of the same
protein samples were separated on one BN-PAGE gel (4–13%) and
immunostained using subtype specific antibodies. (c) Determination of
the oligomeric state of the P2X1+4 heteromer. Purified protein from
oocytes co-injected with equal amounts cRNA encoding His-P2X1 and
P2X4 was partially dissociated by treatment with 100 mM dithiothreitol,
resolved by BN-PAGE, and immunostained with P2X4 and P2X1
antibodies. Numbered arrows indicate the trimeric complex and dissociated monomers and dimers.
P2X4 detected a weaker additional band, most probably
corresponding to a hexameric complex in the non-dissociated
sample (indicated by an arrow in Fig. 1b), suggesting that
P2X4 receptors have a tendency to associate into higher order
complexes, at least if depleted of their natural membrane
environment.
To exclude the possibility that the co-purification was
caused by unspecific aggregation of the over-expressed P2X
subunits, we subjected the purified protein complexes to
BN-PAGE and western blot analysis with specific P2X1 and
P2X4 antibodies. Both P2X1 and P2X4 antibodies labeled a
dominant band of the same size, indicating that both subunits
were present in the same receptor complex (Fig. 1b). The
absence of multiple diffuse bands corresponding to aggregated protein further confirmed that the receptors were
purified in form of heteromeric complexes. Partial dissociation of the receptor complex using dithiothreitol (Nicke
et al. 1998) revealed two additional bands corresponding to
dimers and single subunits, thus demonstrating the trimeric
structure of the complex (Fig. 1c). The antibody against
P2X1 and P2X4 subunits form heteromeric receptors with
unique functional properties
To assess the functional properties of heteromeric P2X1+4
receptors, we took advantage of the distinct biophysical and
pharmacological characteristics of homomeric P2X1 and
P2X4 receptors. P2X1 receptors are fast desensitizing,
activated by abmeATP and highly sensitive to the antagonists suramin and TNP-ATP, whereas P2X4 receptors are
slowly desensitizing, rather insensitive to both abmeATP and
to antagonists (North 2002). In Xenopus oocytes injected
with both subunits, the first application of 10 lM abmeATP
for 5–10 s elicited a mixed response composed of a fast
rising fast desensitizing component followed by a slowly
desensitizing current (Fig. 2a). Subsequent applications of
abmeATP at an interval of 2 min evoked only the slowly
desensitizing component of the current (Fig. 2a). The current
obtained after the first application of abmeATP amounted to
75.7 ± 6.5% (n ¼ 16) of the initially obtained peak current.
In oocytes expressing P2X1 subunits alone, only the fast
rising fast desensitizing current could be observed (Fig. 2b).
Subsequent abmeATP applications at 2-min intervals produced progressively smaller currents (Fig. 2b), with the
second evoked current reduced to 43.5 ± 4.4% (n ¼ 4) of
the initial current. The activation and desensitization rates
were unaffected by the decrease of the current amplitude as
has been previously described for P2X1 receptors (Werner
et al. 1996). The slowly desensitizing current obtained in
oocytes co-injected with P2X1 and P2X4 subunits displayed
2005 International Society for Neurochemistry, J. Neurochem. (2005) 92, 925–933
928 A. Nicke et al.
Fig. 2 Biophysical properties of homomeric P2X1 and heteromeric
P2X1+4 receptors. Superimposed currents obtained from oocytes
expressing P2X1 and P2X4 subunits (a) or P2X1 subunits alone (b)
during consecutive applications (2-min interval) of 10 lM a,b-methylene ATP (abmeATP) are shown. The first application of abmeATP to
oocytes expressing P2X1 and P2X4 subunits evoked a biphasic current with a fast and a slowly desensitizing component. The fast
desensitizing component corresponding to homomeric P2X1 receptors
was absent from the second and following applications due to
incomplete recovery between applications.
kinetic properties similar to homomeric P2X4 receptors
(Figs 3b and c) and showed a time course different from the
current obtained in response to ATP and abmeATP in
oocytes injected with P2X1 subunits (Fig. 3a). Thus, the
current left at the end of a 5 s pulse was 64.8 ± 4.7% (n ¼
8), 55.1 ± 3.5% (n ¼ 5) and 17.8 ± 1.4 (n ¼ 5) of the peak
current for P2X1+4 (at 10 lM abmeATP), P2X4 (at 10 lM
ATP) and P2X1 (at 10 lM abmeATP) receptors, respectively.
It is noteworthy that the percentage of desensitization
observed was more variable for P2X1+4 receptors than for
P2X4 receptors, which might reflect the formation of
heteromeric receptors of different subunit composition. In
oocytes injected with P2X4 subunits alone, no significant
response to the application of 10 lM abmeATP (< 1% of
10 lM ATP) could be detected (Fig. 3b). However, increasing the concentration of abmeATP to 25 lM evoked
measurable currents (Fig. 4a). To further define the characteristics of the P2X1+4 heteromeric receptors, concentration–
response curves for oocytes expressing P2X4 alone or both
P2X1 and P2X4 subunits were obtained (Fig. 4b). To
Fig. 3 Co-expression of P2X1 and P2X4 subunits produced a novel
functional phenotype in Xenopus oocytes. (a) Currents elicited by
application of 10 lM ATP or 10 lM a,b-methylene ATP (abmeATP) on
Xenopus oocytes expressing homomeric P2X1 receptors. For both
agonists, a rapidly activating and rapidly desensitizing current was
observed. Currents were consecutively obtained in the same oocyte
after washing for 2 min, thus only a fraction of the expressed P2X1
receptors was activated by 10 lM abmeATP application. (b) In Xenopus oocytes expressing homomeric P2X4 receptors, a slowly
desensitizing ATP-activated current was observed. No measurable
current was detected when the oocytes were challenged with 10 lM
abmeATP. Currents were measured at a 5-min interval to ensure
recovery of the P2X4 receptors. (c) Currents elicited by ATP and
abmeATP in Xenopus oocytes expressing both P2X1 and P2X4 subunits. The response obtained for 10 lM abmeATP resembles the
response obtained for ATP at homomeric P2X4 receptors. Responses
were obtained at a 2-min interval. Under these conditions most of the
homomeric P2X1 receptors are desensitized.
minimize the contribution of homomeric P2X1 receptors to
our analysis of the peak current elicited by abmeATP at
oocytes injected with P2X1 and P2X4 subunits, we applied
2005 International Society for Neurochemistry, J. Neurochem. (2005) 92, 925–933
Heteromeric receptors formed by P2X1 and P2X4 subunits 929
the agonist at a 2-min interval, which allows only partial
recovery of P2X1 receptors (Figs 2 and 4a) and measured the
current 1 s after current onset. Under these conditions, a
stable response for P2X1+4 receptors showing a slower
activation and desensitization kinetics than P2X1 receptors
was observed (Fig. 4a). abmeATP showed an increased
effect at oocytes injected with P2X1 and P2X4 subunits when
compared to oocytes injected with P2X4 alone. Thus the
maximal current evoked by abmeATP was 15% of the
response to 100 lM ATP for oocytes expressing P2X1 and
P2X4, whereas for homomeric P2X4 receptors, the maximal
abmeATP evoked current was 4.7% of the maximal ATP
response (Fig. 4b). The agonist abmeATP was also found to
be significantly more potent at P2X1+4 (EC50 ¼
35.8 ± 6.5 lM, n ¼ 4–11) than at P2X4 receptors (EC50 ¼
81.1 ± 10.4 lM, n ¼ 3–8). As a contribution of homomeric
P2X4 receptors to the peak current evoked by high concentrations of abmeATP in oocytes injected with P2X1 and
P2X4 subunits could not be excluded, a sum of two
concentration–response curves was fitted to the experimental
data (Fig. 4b). The obtained mean EC50 values were 0.9 and
61.0 lM for the first and second part of the curve, respectively, indicating the presence of at least two receptor
populations. The receptor population showing the lower
EC50 value for abmeATP most probably corresponds to
P2X1+4 receptors, as there is no contribution of P2X1
receptors to the peak currents.
We also tested the sensitivity of P2X1+4 receptors to the
antagonists suramin and TNP-ATP (Fig. 5). Co-application
of suramin (10 lM) blocked 85.5 ± 1.5% (n ¼ 7) of the
current elicited by 10 lM abmeATP at P2X1+4 receptors,
whereas 87.3 ± 3.1% (n ¼ 3) of the current was blocked at
P2X1 receptors under the same conditions (Figs 5a and c).
Currents elicited at P2X4 receptors by 10 lM ATP were
slightly potentiated by 10 lM suramin (result not shown) as
previously described (Bo et al. 1995). To compare the
sensitivity to suramin of the currents elicited by abmeATP at
P2X1+4 and P2X4 receptors, higher concentrations of agonist
were used. Thus, at 100 lM abmeATP, co-application of
suramin (10 lM) blocked 53.5 ± 2.1% (n ¼ 7) of the current
in oocytes injected with P2X1 and P2X4 subunits, whereas
only 2.9 ± 1.4% (n ¼ 5) of the current was blocked under
the same conditions at homomeric P2X4 receptors (Figs 5b
and c). The antagonist TNP-ATP (500 nM) blocked
62.2 ± 2.3 (n ¼ 6) of the current elicited by co-application
of 10 lM abmeATP at P2X1+4 receptors, whereas under the
same conditions 77.3 ± 2.9% (n ¼ 4) of the current through
P2X1 receptors was blocked (Figs 5a and c). These data
agree with what has been described for the co-application of
TNP-ATP and abmeATP at P2X1 receptors in Xenopus
oocytes (Le et al. 1999) and indicate a slightly higher
sensitivity of P2X1 receptors than P2X1+4 receptors to the
blockade by TNP-ATP. Currents through homomeric P2X4
receptors are far less sensitive to TNP-ATP showing an
estimated IC50 of 15 lM (Virginio et al. 1998).
Discussion
Fig. 4 Sensitivity of P2X1+4 and P2X4 receptors to a,b-methylene ATP
(abmeATP). (a) Representative currents obtained upon application of
different concentrations of abmeATP on oocytes expressing P2X1+4 or
P2X4 receptors. The agonist was applied for 5 s at 2-min intervals. This
application scheme allowed full recovery of the abmeATP response at
P2X4 receptors, while fast activating currents corresponding to P2X1
receptors were partially desensitized. However, in order to remove a
possible contribution of P2X1 receptors to our analysis, the peak current was measured 1 s after current onset. (b) Concentration–
response curves for abmeATP on oocytes expressing P2X1+4 (d)
compared to P2X4 (j) receptors. A monophasic (continuous line) or a
biphasic (dashed line) concentration–response curve was fitted to the
data by non-linear regression analysis using commercial software (Igor
Pro, v4.05A Carbon, WaveMetrics, Lake Oswego, OR, USA). Data are
mean ± SEM of three to 11 experiments.
In situ hybridization and immunohistochemistry experiments
have shown that P2X4 is the most widely distributed P2X
subunit (Buell et al. 1996; Collo et al. 1996; Soto et al.
1996; Le et al. 1998b; Bo et al. 2003). However, native P2X
receptors with functional properties similar to heterologously
expressed homomeric P2X4 receptors are scarce. One
possible reason for this discrepancy is the heteromerization
of P2X4 with other P2X subunits. For instance, P2X4
subunits heteromerize with P2X6 subunits in both Xenopus
oocytes (Le et al. 1998a; Khakh et al. 1999) and HEK cells
(Torres et al. 1999). Nevertheless, the functional phenotype
observed presents only minor differences when compared to
homomeric P2X4 receptors (North 2002). As P2X1 subunits
show overlapping distribution with P2X4 subunits in several
tissues, we investigated whether P2X1 and P2X4 subunits
form heteromeric receptors using both biochemical and
electrophysiological methods.
To study the interaction between P2X1 and P2X4 subunits,
a co-purification protocol using a nickel-binding resin was
used. In co-injected Xenopus oocytes, P2X4 subunits were
2005 International Society for Neurochemistry, J. Neurochem. (2005) 92, 925–933
930 A. Nicke et al.
Fig. 5 Sensitivity of P2X1+4 receptors to antagonists. (a) Inhibition of
the current elicited by 10 lM a,b-methylene ATP (abmeATP) by coapplication of 10 lM suramin or 500 nM 2¢,3¢-O-(2,4,6-trinitrophenyl)ATP (TNP-ATP) on P2X1+4 (upper traces) and P2X1 receptors (lower
traces). (b) Inhibition of the current elicited by 100 lM abmeATP by coapplication of 10 lM suramin on oocytes expressing P2X1+4 (upper
traces) or P2X4 (lower traces) receptors. (c) Bar diagrams representing
the amount of current blocked by the antagonists suramin and TNP
(n ¼ 4–7) at P2X1+4 and P2X1 receptors (upper graphic) and by 10 and
100 lM suramin (n ¼ 3–7) at P2X1+4 and P2X4 receptors (lower graphic). Data are mean ± SEM of n experiments. Statistical comparison
was made using the unpaired Student’s t-test: significantly different,
*p < 0.02, **p < 0.001.
co-purified with His-tagged P2X1 subunits indicating the
formation of heteromeric complexes. This is in contrast to a
previous study where no significant interaction was found
between P2X4-His and P2X1-Flag tagged subunits expressed
in HEK-cells (Le et al. 1998a). However, in the mentioned
study, the amount of expressed P2X1 protein was significantly lower than that of the co-expressed P2X4 subunit (Le
et al. 1998a), which might have technically complicated the
detection of the heteromeric receptor. In addition, the relative
abundance of P2X subunits has been proposed to direct the
preferential formation of heteromeric or homomeric P2X
receptors (Calvert and Evans 2004) and might also account
for the observed discrepancy.
To differentiate between an unspecific aggregation of
P2X1 and P2X4 subunits or of the respective homomeric
receptors and the specific formation of heteromeric P2X1+4
receptors, we determined the quaternary structure of the
purified complexes using BN-PAGE analysis. We found that
P2X1 and P2X4 subunits coassemble in a trimeric receptor
complex that matches the trimeric structure of P2X1, P2X3
(Nicke et al. 1998), P2X2 and P2X1+2 receptors (Aschrafi
et al. 2004). Previously published data showed that TritonX100 solubilized P2X4 subunits interact and co-purify in an
unspecific manner with other P2X subunits, including P2X1
(Torres et al. 1999). However, under our experimental
conditions (purification under native conditions and use of
digitonin as detergent), P2X4 subunits primarily co-purify in
the form of trimeric complexes, and only a small amount of
higher order complexes (hexameric) was detected. Detergentdependent hexamer formation has previously been described
(Nicke et al. 1998). Since the hexameric complexes were not
detected with the P2X1 specific antibody, we conclude that in
the presence of digitonin, the formation of stable adducts
between homomeric P2X1 and P2X4 receptors did not occur.
In agreement with this, no P2X4 receptors were detected
when the extracts of oocytes separately injected with HisP2X1 or P2X4 subunits were subjected to the purification
procedure (results not shown).
The functional properties of heteromeric P2X1+4 receptors
were investigated by two-electrode voltage-clamp measurements in Xenopus laevis oocytes. This expression system has
been successfully employed to demonstrate the functional
heteromerization of P2X4 and P2X6 (Le et al. 1998a), P2X1
and P2X5 (Le et al. 1999), P2X1 and P2X2 (Brown et al.
2002) and P2X2 and P2X6 (King et al. 2000) subunits. In all
oocytes co-injected with P2X1 and P2X4 subunits, hetero-
2005 International Society for Neurochemistry, J. Neurochem. (2005) 92, 925–933
Heteromeric receptors formed by P2X1 and P2X4 subunits 931
meric P2X1+4 receptors mediated a slowly desensitizing
current with kinetic parameters resembling the current
through homomeric P2X4 receptors. However, unlike homomeric P2X4 receptors, the heteromeric receptors were activated by low concentrations of abmeATP and inhibited by low
concentrations of the antagonists suramin and TNP-ATP. The
variability of the desensitization kinetics and the biphasic
concentration–response curve obtained for abmeATP activated currents in oocytes injected with P2X1 and P2X4 subunits,
suggest that P2X1+4 receptors showing two different stoichiometries exist. The receptors corresponding to the initial
phase of the concentration–response curve have an EC50
value similar to the one described for homomeric P2X1
receptors (North 2002), while showing the slow activation
and desensitization kinetics (Fig. 4a, current traces to
concentrations lower than 10 lM abmeATP) of homomeric
P2X4 receptors. The second part of the curve most probably
reflects a mixed population of heteromeric P2X1+4 and
homomeric P2X4 receptors. We infer that heteromeric P2X1+4
receptors contribute to the second part of the curve, because
the current obtained upon application of 100 lM abmeATP in
oocytes injected with P2X1 and P2X4 subunits is blocked to
50% by 10 lM suramin (Figs 5b and c). Homomeric P2X4
receptors are insensitive to the antagonist at such low
concentration (Fig. 5b, North 2002). Although a fixed
stoichiometry has been proposed for P2X2+3 receptors (Jiang
et al. 2003), heteromeric P2X receptors with different
stoichiometries has been suggested for the P2X1+5 (Le et al.
1999) and the P2X1+2 combination (Brown et al. 2002).
Similar to what has been observed for heteromeric P2X2+3
receptors (Lewis et al. 1995), the kinetic properties of P2X1+4
receptors are dominated by the slow-desensitizing subunit,
whereas the pharmacological properties resemble that of the
fast-desensitizing subunit. The current amplitude obtained by
application of 10 lM abmeATP on oocytes injected with
P2X1+4 subunits was around 6% of that obtained upon
application of 10 lM ATP. We tried to increase the amount of
heteromeric receptor formed by altering the ratio of P2X1 and
P2X4 cRNAs injected. However, this resulted in an increased
number of the respective homomeric P2X1 or P2X4 receptors
(not shown) in a similar way to what has been described for
the P2X1+2 heteromer expressed in Xenopus oocytes (Brown
et al. 2002). The small current amplitude might indicate low
efficiency of the formation of heteromeric receptors but there
are other possible explanations. For example, abmeATP
could be a partial agonist at P2X1+4 receptors or its single
channel conductance might be different from the corresponding homomeric receptors. Alternatively, auxiliary proteins
aiding formation or trafficking of heteromeric receptors might
be absent or insufficiently expressed in the oocyte expression
system. Along these lines, there are several examples of
heterologously expressed P2X receptors showing small
maximal currents (Collo et al. 1996; Garcia-Guzman et al.
1996) or low efficiency of heterologous expression (Collo
et al. 1996; Soto et al. 1996b; Le et al. 1998a; King et al.
2000; Brown et al. 2002; Jones et al. 2004). However,
whether this reflects the situation in native tissues remains to
be determined. Nevertheless, the detection of functionally
distinct P2X1+4 receptors unambiguously confirms the heteromerization observed using biochemical approaches and
validates BN-PAGE analysis as a useful method to discriminate heteromer formation from unspecific aggregation.
Genetic and pharmacological studies have provided evidence for an involvement of homomeric P2X1 receptors in
purinergic transmission in rodent vas deferens (Liang et al.
2000; Mulryan et al. 2000; Vial and Evans 2002). In this
tissue, expression of P2X1 subunits is progressively up-regulated with development (Liang et al. 2001), and regional
differences in the size of the purinergic response have been
found (Knight et al. 2003). Moreover, besides P2X1, also
P2X4 subunits are present in smooth muscle tissue including
the vas deferens and the bladder (Soto et al. 1996a; Vulchanova et al. 1996; Nori et al. 1998; Bo et al. 2003). Evidence
supporting the presence of abmeATP-activated P2X receptors
that are kinetically and pharmacology different to P2X1
receptors, show similar functional behavior to P2X1+4 receptors and are non-uniformly distributed along the vas deferens
has recently been obtained (Brian F. King, personal communication). Furthermore, a relatively sustained abmeATP
activated current has been found in isolated smooth muscle
cells (Friel 1988). However, this might not represent the
situation in whole tissue, as changes in expression levels of P2
receptor subunits in smooth muscle cells in culture have been
described (Erlinge et al. 1998). The absence of functional
P2X receptors both in whole vas deferens and in isolated
smooth muscle cells from P2X1 (–/–) mice (Mulryan et al.
2000) seems to contrast with the data mentioned above and
disagrees with the presence of P2X4 subunit protein and
mRNA in the tissue. However, it is in agreement with the
general observation that native P2X receptors with functional
properties similar to heterologously expressed homomeric
P2X4 receptors are scarce. Recent studies have shown that
P2X4 receptors undergo constitutive and agonist-induced
internalization in neurons (Bobanovic et al. 2002) and the
authors have suggested that in native tissues, P2X4 receptors
are predominantly present in intracellular compartments from
where they are translocated to the plasma membrane only
under certain yet unknown conditions (Bobanovic et al.
2002). In the case of the vas deferens, it might well be that
heteromerization with P2X1 subunits is necessary to stabilize
cell surface expression of P2X4 subunits. However, additional
experiments will be needed to clarify this point.
In the nervous system, functional P2X receptors similar
to P2X1+4 receptors have been found in neurons from
mouse superior cervical ganglia. The majority of P2X
receptor mediated responses in these cells are pharmacologically and kinetically dominated by the P2X2 subunit.
However, the existence of residual P2X responses in
2005 International Society for Neurochemistry, J. Neurochem. (2005) 92, 925–933
932 A. Nicke et al.
P2X2/P2X3 (–/–) mice strongly points to the participation of
additional P2X receptor subunits in the native P2X receptor
population of SCG neurons (Cockayne et al. 2002). Along
these lines, Calvert and Evans (2004) found in 10–15% of
the superior cervical ganglion neurons small abmeATPactivated non-desensitizing currents that were reduced by
80% in P2X1 (–/–) mice. As these currents share some
functional properties with heterologously expressed homomeric P2X1 and P2X2 receptors, the authors proposed that
they were mediated by a heteromeric receptor containing
P2X1 and P2X2 subunits (Calvert and Evans 2004) but did
not exclude the involvement of other subunits. As in
particular the pH sensitivity and the current kinetics of the
abmeATP mediated response in SCG neurons do not
correlate with those of heteromeric P2X1+2 receptors
heterologously expressed in Xenopus oocytes (Brown et al.
2002; Calvert and Evans 2004), we propose that heteromeric P2X1+4 receptors might be mediating the abmeATP
response measured in SCG neurons. In support of this idea,
that will need further experiments to be validated, both
P2X4 subunit mRNA and protein have been detected in
sympathetic neurons (Dunn et al. 2001).
In summary, we provide biochemical and functional
evidence for the formation of heteromeric P2X1+4 receptors
in Xenopus oocytes. P2X1+4 receptors are trimeric receptors with kinetic properties similar to homomeric P2X4
receptors and a pharmacological profile resembling homomeric P2X1 receptors. Further studies are necessary to
determine if these receptors correspond to the abmeATP
activated slowly desensitizing P2X1-containing receptors
found in native tissues.
Acknowledgements
The authors are grateful to Kerstin Borchardt and Benjamin
Marquez-Klaka for cRNA preparation and to the technical staff of
the Neuropharmacology group for oocyte preparation. We are
indebted to G. Schmalzing for providing the rat P2X1 cDNA. We are
very grateful to Brian F. King for sharing data prior to publication.
We thank Walter Stühmer for generous support and Jürgen Rettinger
for helpful discussion. This work was financed by grants from the
German Israeli Foundation to FS and from the Deutsche Forschung
Gemeinschaft to AN.
References
Aschrafi A., Sadtler S., Niculescu C., Rettinger J. and Schmalzing G.
(2004) Trimeric architecture of homomeric P2X (2) and
heteromeric P2X (1+2) receptor subtypes. J. Mol. Biol. 342, 333–
343.
Bo X., Zhang Y., Nassar M., Burnstock G. and Schoepfer R. (1995) A
P2X purinoceptor cDNA conferring a novel pharmacological
profile. FEBS Lett. 375, 129–133.
Bo X., Kim M., Nori S. L., Schoepfer R., Burnstock G. and North R. A.
(2003) Tissue distribution of P2X4 receptors studied with an
ectodomain antibody. Cell Tissue Res. 313, 159–165.
Bobanovic L. K., Royle S. J. and Murrell-Lagnado R. D. (2002) P2X
receptor trafficking in neurons is subunit specific. J. Neurosci. 22,
4814–4824.
Brown S. G., Townsend-Nicholson A., Jacobson K. A., Burnstock G.
and King B. F. (2002) Heteromultimeric P2X (1/2) receptors show
a novel sensitivity to extracellular pH. J. Pharmacol. Exp. Ther.
300, 673–680.
Buell G., Lewis C., Collo G., North R. A. and Surprenant A. (1996) An
antagonist-insensitive P2X receptor expressed in epithelia and
brain. EMBO J. 15, 55–62.
Calvert J. A. and Evans R. J. (2004) Heterogeneity of P2X receptors in
sympathetic neurons: contribution of neuronal P2X1 receptors
revealed using knockout mice. Mol. Pharmacol. 65, 139–148.
Cockayne D. A., Dunn P. M., Burnstock G. and Ford A. P. D. (2002)
Generation and electrophysiological characterization of P2X2 and
P2X2/3 knockout (KO) mice. Soc. Neurosci. Abstr. 52.12.
Collo G., North R. A., Kawashima E., Merlo-Pich E., Neidhart S.,
Surprenant A. and Buell G. (1996) Cloning OF P2X5 and P2X6
receptors and the distribution and properties of an extended family
of ATP-gated ion channels. J. Neurosci. 16, 2495–2507.
Dunn P. M., Zhong Y. and Burnstock G. (2001) P2X receptors in peripheral neurons. Prog. Neurobiol. 65, 107–134.
Erlinge D., Hou M., Webb T. E., Barnard E. A. and Moller S. (1998)
Phenotype changes of the vascular smooth muscle cell regulate P2
receptor expression as measured by quantitative RT-PCR. Biochem.
Biophys. Res. Commun. 248, 864–870.
Friel D. D. (1988) An ATP-sensitive conductance in single smooth
muscle cells from the rat vas deferens. J. Physiol. 401, 361–380.
Garcia-Guzman M., Soto F., Laube B. and Stuhmer W. (1996) Molecular
cloning and functional expression of a novel rat heart P2X purinoceptor. FEBS Lett. 388, 123–127.
Hu B., Senkler C., Yang A., Soto F. and Liang B. T. (2002) P2X4
receptor is a glycosylated cardiac receptor mediating a positive
inotropic response to ATP. J. Biol. Chem. 277, 15 752–15 757.
Jiang L. H., Kim M., Spelta V., Bo X., Surprenant A. and North R. A.
(2003) Subunit arrangement in P2X receptors. J. Neurosci. 23,
8903–8910.
Jones C. A., Vial C., Sellers L. A., Humphrey P. P., Evans R. J. and Chessell
I. P. (2004) Functional regulation of P2X6 receptors by N-linked
glycosylation: identification of a novel {alpha}{beta}-methylene
ATP-sensitive phenotype. Mol. Pharmacol. 65, 979–985.
Khakh B. S., Proctor W. R., Dunwiddie T. V., Labarca C. and Lester H.
A. (1999) Allosteric control of gating and kinetics at P2X(4)
receptor channels. J. Neurosci. 19, 7289–7299.
King B. F., Townsend-Nicholson A., Wildman S. S., Thomas T., Spyer
K. M. and Burnstock G. (2000) Coexpression of rat P2X2
and P2X6 subunits in Xenopus oocytes. J. Neurosci. 20, 4871–
4877.
Knight D., D’Arbe M., Liang S., Phillips W. D. and Lavidis N. A. (2003)
Regional differences in sympathetic purinergic transmission along
the length of the mouse vas deferens. Synapse 47, 225–235.
Le K. T., Babinski K. and Seguela P. (1998a) Central P2X4 and P2X6
channel subunits coassemble into a novel heteromeric ATP
receptor. J. Neurosci. 18, 7152–7159.
Le K. T., Villeneuve P., Ramjaun A. R., McPherson P. S., Beaudet A.
and Seguela P. (1998b) Sensory presynaptic and widespread
somatodendritic immunolocalization of central ionotropic P2X
ATP receptors. Neuroscience 83, 177–190.
Le K. T., Boue-Grabot E., Archambault V. and Seguela P. (1999)
Functional and biochemical evidence for heteromeric ATP-gated
channels composed of P2X1 and P2X5 subunits. J. Biol. Chem.
274, 15 415–15 419.
Lewis C., Neidhart S., Holy C., North R. A., Buell G. and Surprenant A.
(1995) Coexpression of P2X2 and P2X3 receptor subunits can
2005 International Society for Neurochemistry, J. Neurochem. (2005) 92, 925–933
Heteromeric receptors formed by P2X1 and P2X4 subunits 933
account for ATP-gated currents in sensory neurons. Nature 377,
432–435.
Liang S. X., D’Arbe M., Phillips W. D. and Lavidis N. A. (2000)
Development of fast purinergic transmission in the mouse vas
deferens. Synapse 37, 283–291.
Liang S. X., Motin L., Moussa C. E., Lavidis N. A. and Phillips W. D.
(2001) Spatial distribution and developmental appearance of
postjunctional P2X1 receptors on smooth muscle cells of the
mouse vas deferens. Synapse 42, 1–11.
Mulryan K., Gitterman D. P., Lewis C. J. et al. (2000) Reduced vas
deferens contraction and male infertility in mice lacking P2X1
receptors. Nature 403, 86–89.
Naemsch L. N., Weidema A. F., Sims S. M., Underhill T. M. and Dixon
S. J. (1999) P2X (4) purinoceptors mediate an ATP-activated, nonselective cation current in rabbit osteoclasts. J. Cell Sci. 112, 4425–
4435.
Nicke A., Baumert H. G., Rettinger J., Eichele A., Lambrecht G.,
Mutschler E. and Schmalzing G. (1998) P2X1 and P2X3 receptors
form stable trimers: a novel structural motif of ligand-gated ion
channels. EMBO J. 17, 3016–3028.
Nori S., Fumagalli L., Bo X., Bogdanov Y. and Burnstock G. (1998)
Coexpression of mRNAs for P2X1, P2X2 and P2X4 receptors in
rat vascular smooth muscle: an in situ hybridization and RT-PCR
study. J. Vasc. Res. 35, 179–185.
North R. A. (2002) Molecular physiology of P2X receptors. Physiol.
Rev. 82, 1013–1067.
Patel M. K., Khakh B. S. and Henderson G. (2001) Properties of native
P2X receptors in rat trigeminal mesencephalic nucleus neurones:
lack of correlation with known, heterologously expressed P2X
receptors. Neuropharmacology 40, 96–105.
Rubio M. E. and Soto F. (2001) Distinct Localization of P2X receptors
at excitatory postsynaptic specializations. J. Neurosci. 21, 641–
653.
Schägger H. and Von Jagow G. (1991) Blue native electrophoresis for
isolation of membrane protein complexes in enzymatically active
form. Anal. Biochem. 199, 223–231.
Schägger H., Cramer W. A. and Von Jagow G. (1994) Analysis of
molecular masses and oligomeric states of protein complexes by
blue native electrophoresis and isolation of membrane protein
complexes by two-dimensional native gel electrophoresis. Anal.
Biochem. 217, 220–230.
Soto F., Garcia-Guzman M., Gomez-Hernandez J. M., Hollmann M.,
Karschin C. and Stuhmer W. (1996a) P2X4: an ATP-activated
ionotropic receptor cloned from rat brain. Proc. Natl Acad. Sci.
USA 93, 3684–3688.
Soto F., Garcia-Guzman M., Karschin C. and Stuhmer W. (1996b)
Cloning and tissue distribution of a novel P2X receptor from rat
brain. Biochem. Biophys. Res. Commun. 223, 456–466.
Surprenant A., Schneider D. A., Wilson H. L., Galligan J. J. and North R.
A. (2000) Functional properties of heteromeric P2X (1/5) receptors
expressed in HEK cells and excitatory junction potentials in guineapig submucosal arterioles. J. Auton. Nerv. Syst. 81, 249–263.
Torres G. E., Haines W. R., Egan T. M. and Voigt M. M. (1998)
Co-expression of P2X1 and P2X5 receptor subunits reveals a novel
ATP-gated ion channel. Mol. Pharmacol. 54, 989–993.
Torres G. E., Egan T. M. and Voigt M. M. (1999) Hetero-oligomeric
assembly of P2X receptor subunits. Specificities exist with regard
to possible partners. J. Biol. Chem. 274, 6653–6659.
Vial C. and Evans R. J. (2002) P2X(1) receptor-deficient mice establish
the native P2X receptor and a P2Y6-like receptor in arteries. Mol.
Pharmacol. 62, 1438–1445.
Virginio C., Robertson G., Surprenant A. and North R. A. (1998) Trinitrophenyl-substituted nucleotides are potent antagonists selective
for P2X1, P2X3, and heteromeric P2X2/3 receptors. Mol. Pharmacol. 53, 969–973.
Vulchanova L., Arvidsson U., Riedl M., Wang J., Buell G., Surprenant
A., North R. A. and Elde R. (1996) Differential distribution of two
ATP-gated channels (P2X receptors) determined by immunocytochemistry. Proc. Natl Acad. Sci. USA 93, 8063–8067.
Weidema A. F., Barbera J., Dixon S. J. and Sims S. M. (1997) Extracellular nucleotides activate non-selective cation and Ca(2+)dependent K+ channels in rat osteoclasts. J. Physiol. 503, 303–315.
Werner P., Seward E. P., Buell G. N. and North R. A. (1996) Domains of
P2X receptors involved in desensitization. Proc. Natl Acad. Sci.
USA 93, 15 485–15 490.
Zhong Y., Dunn P. M., Bardini M., Ford A. P., Cockayne D. A. and
Burnstock G. (2001) Changes in P2X receptor responses of sensory
neurons from P2X3-deficient mice. Eur. J. Neurosci. 14, 1784–1792.
2005 International Society for Neurochemistry, J. Neurochem. (2005) 92, 925–933